KR20150061914A - Skin texture predicting method and apparatus - Google Patents

Abstract

A method and a device to predict a skin age are disclosed. The disclosed method to predict a skin age comprises steps of: calculating scores of plural parameters related to a skin age of a user by using information to record a life pattern of the user during a past time previously decided at a first point, and estimating the skin age for the user by using the calculated scores of the plural parameters and weighted values of the plural parameters, previously decided; measuring the skin age of the user by using a skin microscope image of the user; updating the weighted values of the plural parameters by using the estimated skin age and the measured skin age; and predicting a skin age of the user after the first point, by using the updated weighted values and the calculated scores of the plural parameters.

Description

[0001] The present invention relates to a skin texture predicting method and apparatus,

Embodiments of the present invention relate to a method and apparatus for predicting skin age using a user's life pattern.

Human skin is the most extrinsic part of the body. It is a part of our body that protects our environment from environmental pollution, bacteria, etc., and provides useful information related to health.

However, as the skin becomes older, symptoms related to aging appear, and the development of cosmetics and procedures to alleviate it is increasing. However, there is no criterion to quantify the condition of your skin, and many people are dependent on the opinion of the dermatologist.

Therefore, it is necessary to establish a quantitative statistic for human skin condition, and to study the method for objectively deriving human skin age based on this.

On the other hand, a service for predicting the skin age is provided through a user device such as a smart phone. However, the skin management software which has been studied so far judges the user's skin by the average threshold value, and does not consider the life pattern of the user causing the problem.

In order to solve the problems of the related art as described above, the present invention proposes a method and apparatus for predicting skin age using a user's life pattern.

Other objects of the invention will be apparent to those skilled in the art from the following examples.

According to a preferred embodiment of the present invention, there is provided a method for recording a user's life pattern for a predetermined time period at a first point of time, Estimating a skin age of the user based on a score of the calculated plurality of parameters and a predetermined weight of the plurality of parameters; Measuring the skin age of the user using the skin microscope image of the user; Updating the weights of the plurality of parameters using the estimated skin age and the measured skin age; And estimating a skin age of the user after the first point of time using the updated weight and the score of the calculated plurality of parameters.

Information for recording the life pattern of the user can be collected through at least one of a social network service (SNS) post, a calendar application, a diary application, and a household application.

The plurality of parameters may include at least one of a food intake parameter, an outdoor activity parameter, a movement parameter, a sleep parameter, a drinking parameter and a smoking parameter.

The score of the food intake parameter may be calculated based on at least one of the user's age, height, weight, calorie of the user's ingested food, and ingestion time of the user's ingested food.

The score of the outdoor activity parameter may be calculated based on at least one of an outdoor temperature, an outdoor activity duration of the user, time information when the user performs an outdoor activity, and cloudiness.

The score of the motion parameter may be calculated based on at least one of the number of motions and the duration of the motions.

The score of the sleep parameter may be calculated based on at least one of a sleep start time and a sleep duration.

The drinking parameter may be calculated based on at least one of a drinking amount of the user and a calorie amount according to the drinking amount.

The step of measuring the skin age of the user comprises the steps of: pre-processing the skin micrographic image of the user and converting the image into a skeleton image; Detecting each skin cell through polygon mesh detection in the skeleton image; Extracting feature information of each detected skin cell, the feature information including the skin cell average size and number; And deriving skin age using feature information of the skin cell.

According to another embodiment of the present invention, there is provided a method for calculating a score of a plurality of parameters associated with a skin age of a user using information for recording a user's life pattern for a predetermined past time at a first time point, An estimating unit that estimates a skin age of the user based on a score of the plurality of calculated parameters and a weight of the predetermined plurality of parameters; A measuring unit measuring the skin age of the user using the skin micrograph image of the user; An updating unit updating the weights of the plurality of parameters using the estimated skin age and the measured skin age; And a predictor for predicting a skin age of the user after the first point of time using the updated weight and the score of the calculated plurality of parameters.

According to the present invention, the skin age can be predicted using the life pattern of the user.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic view showing a skin age predicting apparatus according to an embodiment of the present invention; FIG.
2 is a flowchart illustrating a skin age prediction method according to an embodiment of the present invention.
FIG. 3 illustrates the concept of the FQL (Facebook Query Language). FIG.
FIG. 4 is a flowchart illustrating a method of measuring skin age using a skin micrograph image according to an embodiment of the present invention. FIG.
5 is a flowchart illustrating a method of detecting a skin cell according to an embodiment of the present invention.
6 is a diagram illustrating pseudo code for skin cell detection according to an embodiment of the present invention.
7 illustrates a detected skin cell according to one embodiment of the present invention.
FIG. 8 is a view for explaining an over-divided skin cell according to an embodiment of the present invention; FIG.
FIG. 9 is a view for explaining merging of over-divided skin cells according to an embodiment of the present invention; FIG.
10 is a flowchart illustrating a method of extracting feature information of a skin cell according to an embodiment of the present invention.
11 is a flowchart for explaining a method of deriving an inscribed circle / circumference ratio according to an embodiment of the present invention.
12 illustrates a detected inscribed circle and a circumscribed circle according to an embodiment of the present invention.
13 is a view for explaining an intersection detection according to an embodiment of the present invention;
Figure 14 illustrates cross-point connections in accordance with one embodiment of the present invention.
FIGS. 15 to 17 are diagrams illustrating statistics of feature information of skin cells by age according to an embodiment of the present invention; FIG.
18 is a view schematically showing an internal configuration of a skin age measuring apparatus according to an embodiment of the present invention.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. Like reference numerals are used for like elements in describing each drawing.

Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a view showing a schematic configuration of a skin age predicting apparatus according to an embodiment of the present invention.

1, the apparatus 100 for predicting skin age according to an embodiment of the present invention may be a smart device such as a smart phone, a tablet PC, etc., and may include a collection unit 110, an estimation unit 120, 130, an updating unit 140, and a predicting unit 150.

2 is a flowchart illustrating a skin age prediction method according to an embodiment of the present invention.

Hereinafter, the function of each component of the skin age predicting apparatus 100 and the steps of the skin age predicting method will be described with reference to FIGS. 1 and 2, respectively.

First, in step S210, the collecting unit 110 collects information for recording a user's life pattern for a predetermined past time at a first time point.

Here, the first viewpoint may be the "current viewpoint ", and the predetermined past time may be" one week ". Hereinafter, for convenience of explanation, the first viewpoint will be referred to as "current viewpoint ".

According to an embodiment of the present invention, the collecting unit 110 may collect information for recording a user's life pattern through at least one of a social network service (SNS) post, a calendar application, a diary application, have.

For example, the collecting unit 110 may collect information for recording a life pattern of a user using a Facebook post (timeline). In this case, the collecting unit 110 may collect the information (e.g., the place visited by the user, the time (including the check-in time), the purpose, etc.) using the FQL (Facebook Query Language) .

Next, in step S215, the estimator 120 calculates the score of the plurality of parameters associated with the skin age of the user using the information for recording the life pattern of the user.

According to one embodiment of the present invention, the plurality of parameters may include at least one of a food intake parameter, an outdoor activity parameter, an exercise parameter, a sleep parameter, a drinking parameter and a smoking parameter.

The contents of each parameter will be described in detail below.

A. Food intake parameter

In the case of the food intake parameter, the collecting unit 110 collects information on the age, the height, the weight, the calorie of the food consumed by the user, the consumption time of the food consumed by the user (when food is consumed) And the estimating unit 120 calculates the score of the food intake parameter using the above information.

In addition, the estimating unit 120 may use a known standard calorie amount for the food to calculate the calorie of the food consumed by the user, and the collecting unit 110 may collect the standard calories through the communication network .

According to an embodiment of the present invention, the estimating unit 120 may calculate the score of the food intake parameter based on the following equation (1).

Here, DPW _eating scores of food intake parameters, BM is the primary metabolism, which is determined by your age, height, weight (Basic Metabolism), Actual intake is calorimetry of the food that you _consume, W _{e, timezone} is food intake Time information, and min () denotes a minimum function, respectively. The basic metabolism can be defined based on the following equation (2).

In addition, the user consuming food at night late results in bad skin results, which is reflected in the W _e , _timezone of food intake in equation (1). The W _e , _timezone of food intake can be defined as shown in Table 1.

Morning Afternoon Evening Midnight W _e , _timezone One One One 1.5

In summary, the score of the food intake parameter has a value between 0 and 1 by the minimum function. If the caloric value of the user's basic metabolism is similar to that of the user's basic metabolism, the score of the food intake parameter is close to zero, and if the user consumes too much food or consumes too little food, The score has a value close to 1.

N. Outdoor activity parameter

In the case of the outdoor activity parameter, the collection unit 110 collects outdoor temperature, user's outdoor activity duration, time information when the user performs outdoor activity (time of outdoor activity), cloudiness And the estimating unit 120 calculates the score of the outdoor activity parameter using the above information.

According to an embodiment of the present invention, the estimation unit 120 may calculate the score of the outdoor activity parameter based on the following equation (3).

Where DPW _outdoor is the outdoor temperature parameter, T _duration is the outdoor temperature when the temperature is lower than t ₁ ° C (eg, 5 ° C) or t ₂ ° C (eg, 27 ° C) T _outdoor is the outdoor activity duration when t ₁ ° C or higher than t ₂ ° C during the day, W _{o, timezone} is the time information when the user is outdoors, UV _cloud is the day cloudiness, W _{UV_expo} is a value obtained by scaling the day cloudiness, α and β are weights (α = 0.3, β = 0.7), and max () is a maxima function.

In addition, W _{o , timezone} when a user performs an outdoor activity can be defined as shown in Table 2.

Morning Afternoon Evening Midnight W _{o , timezone} 0.8 1.5 One One

In sum, the score of the outdoor activity parameter is between 0 and 1 by α, β and maxima function, and the user's outdoor activity is almost zero or the outdoor temperature is higher than t ₁ ° C or t ₂ ° C lower , The score of the outdoor activity parameter has a value close to 0, and the score of the outdoor activity parameter is close to _{1 in a} case where the outdoor activity of the user is large or the outdoor temperature is lower than t ₁ ° C or t ₂ ° C .

C. Motion parameter

In the case of the exercise parameters, the collection unit 110 collects at least one of the number of exercises and the exercise duration through the SNS, the communication network, and the like, and the estimation unit 120 calculates the score of the exercise parameter using the above information.

According to an embodiment of the present invention, the estimation unit 120 may calculate the score of the motion parameter based on the following equation (4).

Here, DPW _exer scores of motion parameters, DPW _{ex, init} is the initial value of the movement parameter score, total exercise time during T _exer is a predetermined past time, T _{ex, Rec} is recommended exercise time, T _Over excessive And a threshold value of the exercise time classified by the exercise.

In summary, the score of the exercise parameter may have a value between 0 and 1 based on the threshold of the total exercise time, the recommended exercise time and the exercise time classified as excessive exercise over a predetermined past time.

D. Sleep parameters

In the case of the sleep parameter, the collecting unit 110 collects at least one of the sleep start time and the sleep duration through the SNS, the communication network, and the like, and the estimator 120 calculates the score of the sleep parameter .

According to an embodiment of the present invention, the estimator 120 may calculate the score of the sleep parameter based on the following equation (5).

In this case, DPW _sleeping is the score of the sleep parameter, T _sleeping is the sleep duration of the user, T _{s , Rec} is the recommended sleeping time, T _{Ref _ MASRT} is the sleep interval between the first time and the second time emitter being 2 times the period of the most active skin regeneration between the hours), T _{user _ MASRT} is the inclusion of a second time from one hour of your sleep duration (for example, user 23 pm and taken to the surface If the next day was up at 8, T _{User _ MASRT} means 3 (= 11-2) Im), respectively.

Taken together, the score of the sleep parameter can be normalized to a value between 0 and 1 based on at least one of the sleep start time and the sleep duration.

ㅁ. Drinking parameter

In the case of the drinking parameter, at least one of the drinking quantity and the calorie quantity according to the drinking quantity of the user is collected through the SNS, the communication network, etc., and the estimating unit 120 calculates the score of the drinking parameter using the above information.

According to an embodiment of the present invention, the estimating unit 120 may calculate the score of the drinking parameter based on Equation (6) below.

In this case, DPW _drinking is the score of the drinking parameter, DPW _{d , init} is the initial value of the drinking parameter score, C _Thresh is the amount of _drinking allowed per day (for example, 200 calories), C _drinking is the amount of _drinking One value, W _{d , and timezone} , respectively, represent time information when the user drinks alcohol.

In addition, W _{d and timezone} at the time of drinking by the user can be defined as shown in Table 3.

Morning Afternoon Evening Midnight W _{d , timezone} 1.5 One One 1.5

_{Taken together} , the score of the drinking parameter can be normalized to a value between 0 and 1 based on C _Thresh , C _drinking , and W _{d , timezone} . .

F. Smoking parameter

In the case of the smoking parameter, the smoking information is collected through the SNS, the communication network, and the like, and the estimating unit 120 calculates the score of the smoking parameter using the above information. Here, the score of the smoking parameter has a value of 1 when the user smokes, and the score of the smoking parameter has a value of 0 when the user does not smoke.

Subsequently, in step S220, the estimating unit 120 estimates the skin age of the user by using the score of the calculated plurality of parameters and the weight of a predetermined plurality of parameters.

That is, in order to estimate the age of the user, the score of a plurality of parameters is important, but how the plurality of parameters affect the user is an important factor. For example, if the skin age is young, even though a particular user has a lot of alcohol, the score of the drinking parameter does not have a significant effect on the skin age of a particular user. This is due to the genetic characteristics of the user.

Accordingly, in accordance with an embodiment of the present invention, the estimator 120 may calculate a score in which a weight is reflected by using a multiplication operation of a score of a parameter and a weight for each of a plurality of parameters, The skin age of the user can be estimated.

According to one embodiment of the present invention, the initial values of the respective weights of the plurality of parameters may all have the same value, and the values of the weights of the plurality of parameters are adjusted in step S230 described below.

Next, in step S225, the measuring unit 130 measures the age of the user's skin using the user's skin micrograph image. That is, in step S208, the age of the user's skin is directly measured using a skin age determination algorithm. A detailed description of the skin age determination algorithm of the measurement unit 130 will be described later.

Thereafter, in step S230, the updating unit 140 updates the weights of the plurality of parameters using the estimated skin age measured in step S220 and the measured skin age measured in step S225.

That is, in step S230, the initial values of the weight values of the plurality of parameters are updated to reduce the error between the estimated skin age and the measured skin age, Age prediction can be performed.

For example, as described above, when the weight of a plurality of parameters is large, assuming that the age of the skin is old, when the measured skin age is young despite a lot of drinking by the user, The update unit 140 can adjust the weight of the initial drinking parameter downward. On the other hand, when it is determined that the skin age is much older than the other smokers even though the user smokes, the smoker is analyzed to have an absolute influence on the skin age of the user, The weight of the initial smoking parameter can be adjusted upward.

According to an embodiment of the present invention, the update unit 140 may update the weights of a plurality of parameters using a hidden markov model (HMM).

), 상태 i에서 j로의 전이 분포(

) 및 방출 또는 관찰 모델(emission or observation model)(

)로써 표현되어 진다. More specifically, the HMM performs a Markov model representing the type of state transition. The HMM measures the start value (t = 0), the probability distribution

), The transition distribution from state i to j (

) And an emission or observation model (

).

Meanwhile, since the Hidden Markov model is a technique known in the art, a detailed description thereof will be omitted.

Finally, in step S235, the predicting unit 150 predicts the skin age of the user after the first point of time by using the updated weight and the score of the calculated plurality of parameters. Here, the prediction of the prediction unit 150 is applied in the same manner as the estimation operation of the estimation unit 120 described above.

On the other hand, after the skin age is predicted, the skin age prediction apparatus 100 according to the present invention repeatedly performs steps S210 to S230 to further analyze the genetic characteristics of the user, Can be renewed, thereby enabling more precise skin age prediction. Here, the predicted value of the predicting unit 150 may be an estimated value of the following estimating unit 120. That is, the estimator 120 and the estimator 150 may be implemented as a single module.

Hereinafter, the skin age determination algorithm of the measurement unit 130 will be described in detail.

The measuring unit 130 detects each cell constituting the skin after preprocessing the skin micro-microscope image and measures the cell characteristics such as the number of detected cells, average cell width, inscribed circle / circumference ratio, number of intersections, After extraction, the skin age can be derived quantitatively based on this.

FIG. 4 is a flowchart illustrating a method of measuring skin age using a skin micrograph image according to an embodiment of the present invention.

In step S410, the measurement unit 130 preprocesses the skin micro-microscope image.

The photographed skin microscope image is difficult to obtain uniform image quality in the whole area due to limitations of imaging equipment or light source, and vignetting phenomenon occurs. For this reason, vignetting of a skin micrograph image is eliminated by gradation masking on a skin micrograph image, and a center portion is sampled. Because the skin microscope images are acquired in various environments, they have different brightness due to the influence of the ambient light amount, and the histogram leveling process can be performed to improve the contrast of the images.

Next, the noise can be removed by performing a water-sheath algorithm after binarizing by Otsu's method to remove noise from a skin micro-image.

Since the Otsu method and the water-set algorithm are obvious to those skilled in the art, a detailed description thereof will be omitted.

In step S415, the measurement unit 130 divides the area of the binarized image.

For example, the measurement unit 130 displays a binarized gray image through a water-set algorithm as a skeleton image that sets a boundary line of a region to one pixel.

In step S420, the measuring unit 130 detects each cell in the skeleton image.

For example, the measuring unit 130 may detect a component (polygon) connected to four neighboring boundary lines (for example, a left boundary line, a right boundary line, an upper boundary line, and a lower boundary line) Respectively, and set the searched components as skin cells, respectively.

This will be described in more detail with reference to FIG. 5 below.

In step S425, the measurement unit 130 extracts feature information on each detected skin cell.

For example, the measuring unit 130 may extract at least one of the average cell size, the number of skin cells, the inscribed circle / circumference ratio, the number of crossing points, and the crossing average length for each detected skin cell as feature information. This will be described in more detail with reference to FIG. 9 below.

In step S430, the measuring unit 130 deduces the skin age using each extracted skin cell feature information.

Generally, as the skin age of a person increases, the area (size) of the skin cell gradually increases, and the number of skin cells tends to decrease gradually. In addition, the ratio of the inscribed circle to the circumscribed circle (hereinafter referred to as inscribed circle / circumscribed circle) of the skin cells gradually decreases, and the average length between the crossed points tends to increase gradually. Accordingly, the present invention extracts feature information of a skin cell after each skin cell is detected through a skin microscope image, and can deduce skin age based on the feature information.

For example, as shown in FIGS. 15 to 17, skin age information can be inferred by comparing skin cell characteristic information of a subject with reference to statistical values of skin cell characteristic information for each age group.

FIG. 5 is a flowchart illustrating a method of detecting a skin cell according to an embodiment of the present invention. FIG. 6 is a diagram illustrating pseudo code for skin cell detection according to an embodiment of the present invention. FIG. 8 is a view for explaining an over-divided skin cell according to an embodiment of the present invention, and FIG. 9 is a view showing a skin cell according to an embodiment of the present invention FIG. 5 is a view for explaining merging of over-divided skin cells according to FIG.

In step S510, the measuring unit 130 detects the polygons connected to at least four neighboring boundary lines in the converted skeleton image, respectively. The thus detected polygon will be collectively referred to as a skin cell.

In step S515, the measuring unit 130 determines whether the skin cell detection process is completed for all the regions of the skeleton image.

That is, the measuring unit 130 may repeatedly perform a process of detecting a polygon connected to at least four boundary lines for all areas of the skeleton image.

FIG. 6 illustrates a pseudo code for detecting a polygon connected to at least four neighboring boundary lines in a skeleton image as a skin cell. 7, the detected skin cells are illustrated.

If detection for all the areas is not completed, the process proceeds to step S510.

However, when the detection of all the areas is completed, the measuring unit 130 calculates the average area size of the skin cells detected in step S520. It goes without saying that the measuring unit 130 may also store the number of detected skin cells.

Then, in step S525, the measurement unit 130 detects the over-divided skin cells.

For example, the measurement unit 130 sets the outline of each cell by applying a convex hull to each cell. The measurement unit 130 may detect the skin cells as over-divided cells when any skin cells having their own center points exist in different skin cells with respect to each of the skin cells to which the convex holes are applied.

FIG. 8 shows an example in which a skin cell over-divided is searched using an outline of each cell retrieved by applying a convex hole to each skin cell. Referring to FIG. 8, the skin cells 810 and 815 each have a center point, but are skin cells contained in the skin 801 and are searched for over-divided skin cells. In addition, the 820 skin cells also have center points, but they are searched for skin cells that are over-divided into skin cells contained within 802.

When the over-divided skin cell search is completed, the measuring unit 130 merges the over-divided skin cells into other skin cells in step S530. For example, if the overdivided skin cells are less than the average area size, then the overdivided skin cells may have different skin cells (e.g., center points of the corresponding cells) Other skin cells containing)). FIG. 9 shows an example in which the over-divided skin cells of FIG. 6 are merged.

When the merging process of the over-divided skin cells is completed, the measuring unit 130 updates the average area size and skin cell number of the skin cells again in step S535.

5, steps S530 to S535 have been performed on the assumption that there are over-divided skin cells, but if there are no over-divided skin cells, steps S530 to S535 are not included It is natural that it may not.

FIG. 10 is a flowchart illustrating a method of extracting feature information of a skin cell according to an embodiment of the present invention. FIG. 12 is a view for explaining an intersection detection according to an embodiment of the present invention, FIG. 3 illustrates a cross-point connection according to an embodiment of the present invention. FIG.

In step S1010, the measuring unit 130 derives the average area size of the detected skin cells and the number of skin cells as feature information.

Next, in step S1015, the measuring unit 130 derives the inscribed circle / circumscribed circle ratio of the detected skin cells. For example, the measuring unit 130 may gradually increase the morphological region with respect to the center point of the skin cell detected to set the inscribed circle and the circumscribed circle, and then derive the inscribed circle / the circumscribed circle ratio. This will be described in more detail below with reference to FIG.

In step S1020, the measurement unit 130 derives the number of crossing points and the average length between the crossing points as feature information through detection of the intersection points of the detected skin cells.

For example, the measuring unit 130 can detect points that meet at least three boundary lines at intersections of the points located in the boundary lines of each detected skin cell, respectively.

FIG. 13 illustrates intersection points detected with respect to arbitrary skin cells. Referring to FIG. 13, it can be seen that 1310a to 1310f are detected as intersections, respectively, at points where they meet at least three boundary lines.

Then, the measuring unit 130 can derive the number of intersections of each skin cell and the average length between the intersections. At this time, the measuring unit 130 may calculate the distance between each intersection by connecting the straight distances between arbitrary intersections and other intersections, and then calculate the average distance by averaging the distances. Since the method of calculating the straight line distance between two points is already obvious to those skilled in the art, a detailed description thereof will be omitted.

FIG. 14 shows an example in which the intersections detected for each skin cell are connected to each other.

In this way, the measuring unit 130 extracts various feature information on the detected skin cells, and can deduce skin age using the extracted feature information.

FIG. 11 is a flowchart illustrating a method of deriving an inscribed circle / circumference ratio according to an embodiment of the present invention, and FIG. 12 is a diagram illustrating detected inscribed circle and circumscribed circle according to an embodiment of the present invention.

In step S1110, the measurement unit 130 gradually expands a morphological region (e.g., a circle) based on the center point of each skin cell.

In step S1115, the measuring unit 130 gradually enlarges the morphological region and sets a prototype to the morphological region up to the point of first meeting with the boundary line of the skin cell (i.e., the point of first meeting among the boundary lines) Set as inscribed circle.

Then, the measuring unit 130 calculates the diameter of the set inscribed circle.

Then, in step S1120, the measuring unit 130 gradually expands the morphological region to a morphological region up to a point at which the boundary line of the skin cell is the last to be encountered Set a circle and set it as a circumscribed circle. Next, the measuring unit 130 calculates the diameter of the set circumscribed circle.

12 shows an example of setting the inscribed circle and the circumscribed circle to the skin cells. In Fig. 12, the bold line represents the inscribed circle, and the thin line represents the circumscribed circle.

In step S1120, the measuring unit 130 calculates the inscribed circle / circumscribed circle ratio using the calculated inscribed circle diameter and the circumscribed circle diameter.

At this time, the measuring unit 130 extracts the top 15 inscribed circle / circumscribed circle ratios among the calculated inscribed circle / circumscribed circle ratios, and can use it as feature information of the skin cell.

15 to 17 are views showing statistical values of feature information of skin cells by age according to an embodiment of the present invention.

FIGS. 15 to 17 are tables showing characteristic data of the skin cells extracted from the skin micrographs of the respective ages and then statistical analysis. Thus, the feature information of the skin cells for each age group is statistically stored and then the feature information of the skin cells can be extracted through the skin microscope image of the subject to objectively deduce the skin age.

FIG. 15 is a graph showing the feature information of the skin cell extracted using statistical images of facial skin microscopes at various ages. FIG. 16 is a graph showing the characteristics of skin cells extracted from statistical data FIG. 17 is a graph showing the extraction of feature information of skin cells using a hand skin microscope image of each age group and statistical analysis.

In the embodiment of the present invention, the skin age is inferred using only the feature information of each skin cell. However, as shown in FIGS. 15 to 17, information such as average length, width and height of the wrinkles May be extracted as feature information and used together with the feature information of the skin cell.

FIG. 18 is a view schematically showing an internal configuration of a skin age measuring apparatus according to an embodiment of the present invention.

18, a measuring unit 130 according to an embodiment of the present invention includes a preprocessing unit 1810, a detecting unit 1815, a feature extracting unit 1820, a reasoning unit 1825, a memory 1830, (1835).

The preprocessing unit 1810 binarizes the skin micrograph image using the Otsu method to remove noise from the skin micrograph image, and then performs a water-sheath algorithm to remove the noise.

The preprocessed image can be converted into a skeleton image that sets the boundary line of the region to one pixel in the gray image binarized by the water-seek algorithm.

The detection unit 1815 is means for detecting skin cells in the skeleton image.

For example, the measuring unit 130 may detect a component (polygon) connected to four neighboring boundary lines (for example, a left boundary line, a right boundary line, an upper boundary line, and a lower boundary line) And each retrieved component can be set as one cell. This is the same as that described above with reference to FIG. 4, so duplicate descriptions will be omitted.

The feature extraction unit 1820 extracts feature information about each detected skin cell.

For example, the measuring unit 130 may extract at least one of the average cell size, the number of skin cells, the inscribed circle / circumference ratio, the number of crossing points, and the crossing average length for each detected skin cell as feature information. Since this is the same as that described with reference to FIG. 7, a duplicate description will be omitted.

The inference unit 1825 deduces the skin age using each extracted skin cell feature information.

For example, as shown in FIGS. 15 to 17, the inferring unit 1825 can infer the skin age by comparing statistical statistics of the skin cell feature information of each age group with the feature information of each skin cell of the subject.

The memory 1830 stores various algorithms required for performing the skin age measuring method according to an embodiment of the present invention, statistics of feature information of each age skin cell, and the like.

The control unit 1835 controls the internal components (for example, the preprocessing unit 1810, the detection unit 1815, the feature extraction unit 1820, the reasoning unit 1825, and the like) of the measurement unit 130 according to an embodiment of the present invention. ), The memory 1830, and the like).

In addition, the above-described technical features may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware device may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, the present invention has been described with reference to particular embodiments, such as specific elements, and limited embodiments and drawings. However, it should be understood that the present invention is not limited to the above- Various modifications and variations may be made thereto by those skilled in the art to which the present invention pertains. Accordingly, the spirit of the present invention should not be construed as being limited to the embodiments described, and all of the equivalents or equivalents of the claims, as well as the following claims, belong to the scope of the present invention .

Claims (10)

Calculating a score of a plurality of parameters associated with the user's skin age using information for recording a user's life pattern for a predetermined past time at a first point of time, Estimating a skin age of the user using a weight of a plurality of parameters;
Measuring the skin age of the user using the skin microscope image of the user;
Updating the weights of the plurality of parameters using the estimated skin age and the measured skin age; And
And estimating the skin age of the user after the first point of time using the updated weight and the score of the calculated plurality of parameters. The method according to claim 1,
Wherein information for recording the life pattern of the user is collected through at least one of a social network service (SNS) post, a calendar application, a diary application, and a household application. The method according to claim 1,
Wherein the plurality of parameters comprises at least one of a food intake parameter, an outdoor activity parameter, an exercise parameter, a sleep parameter, a drinking parameter and a smoking parameter. The method of claim 3,
Wherein the score of the food intake parameter is calculated based on at least one of a user's age, a height, a weight, a calorie of the food consumed by the user, and an intake time of the food consumed by the user. The method of claim 3,
Wherein the score of the outdoor activity parameter is calculated based on at least one of an outdoor temperature, an outdoor activity duration of the user, time information when the user performs an outdoor activity, and cloudiness. The method of claim 3,
Wherein the score of the exercise parameter is calculated based on at least one of a number of times of exercise and a duration of exercise. The method of claim 3,
Wherein the score of the sleep parameter is calculated based on at least one of a sleep start time and a sleep duration. The method of claim 3,
Wherein the drinking parameter is calculated based on at least one of a drinking amount of the user and a calorie amount corresponding to the drinking amount. The method according to claim 1,
Wherein the measuring the skin age of the user comprises:
Transforming the user's skin micro-scope image into a skeleton image by preprocessing;
Detecting each skin cell through polygon mesh detection in the skeleton image;
Extracting feature information of each detected skin cell, the feature information including the skin cell average size and number; And
And deriving a skin age using feature information of the skin cell. Calculating a score of a plurality of parameters associated with the user's skin age using information for recording a user's life pattern for a predetermined past time at a first point of time, An estimating unit estimating a skin age of the user using a weight of a plurality of parameters;
A measuring unit measuring the skin age of the user using the skin micrograph image of the user;
An updating unit updating the weights of the plurality of parameters using the estimated skin age and the measured skin age; And
And a predictor for predicting the skin age of the user after the first point of time using the updated weight and the score of the calculated plurality of parameters.

Images